From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S966389AbeCAG2C (ORCPT ); Thu, 1 Mar 2018 01:28:02 -0500 Received: from mga03.intel.com ([134.134.136.65]:12103 "EHLO mga03.intel.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S966378AbeCAG14 (ORCPT ); Thu, 1 Mar 2018 01:27:56 -0500 X-Amp-Result: SKIPPED(no attachment in message) X-Amp-File-Uploaded: False X-ExtLoop1: 1 X-IronPort-AV: E=Sophos;i="5.47,407,1515484800"; d="scan'208";a="33761289" From: Aaron Lu To: linux-mm@kvack.org, linux-kernel@vger.kernel.org Cc: Andrew Morton , Huang Ying , Dave Hansen , Kemi Wang , Tim Chen , Andi Kleen , Michal Hocko , Vlastimil Babka , Mel Gorman , Matthew Wilcox , David Rientjes Subject: [PATCH v4 3/3] mm/free_pcppages_bulk: prefetch buddy while not holding lock Date: Thu, 1 Mar 2018 14:28:45 +0800 Message-Id: <20180301062845.26038-4-aaron.lu@intel.com> X-Mailer: git-send-email 2.14.3 In-Reply-To: <20180301062845.26038-1-aaron.lu@intel.com> References: <20180301062845.26038-1-aaron.lu@intel.com> Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org When a page is freed back to the global pool, its buddy will be checked to see if it's possible to do a merge. This requires accessing buddy's page structure and that access could take a long time if it's cache cold. This patch adds a prefetch to the to-be-freed page's buddy outside of zone->lock in hope of accessing buddy's page structure later under zone->lock will be faster. Since we *always* do buddy merging and check an order-0 page's buddy to try to merge it when it goes into the main allocator, the cacheline will always come in, i.e. the prefetched data will never be unused. In the meantime, there are two concerns: 1 the prefetch could potentially evict existing cachelines, especially for L1D cache since it is not huge; 2 there is some additional instruction overhead, namely calculating buddy pfn twice. For 1, it's hard to say, this microbenchmark though shows good result but the actual benefit of this patch will be workload/CPU dependant; For 2, since the calculation is a XOR on two local variables, it's expected in many cases that cycles spent will be offset by reduced memory latency later. This is especially true for NUMA machines where multiple CPUs are contending on zone->lock and the most time consuming part under zone->lock is the wait of 'struct page' cacheline of the to-be-freed pages and their buddies. Test with will-it-scale/page_fault1 full load: kernel Broadwell(2S) Skylake(2S) Broadwell(4S) Skylake(4S) v4.16-rc2+ 9034215 7971818 13667135 15677465 patch2/3 9536374 +5.6% 8314710 +4.3% 14070408 +3.0% 16675866 +6.4% this patch 10338868 +8.4% 8544477 +2.8% 14839808 +5.5% 17155464 +2.9% Note: this patch's performance improvement percent is against patch2/3. [changelog stole from Dave Hansen and Mel Gorman's comments] https://lkml.org/lkml/2018/1/24/551 Suggested-by: Ying Huang Signed-off-by: Aaron Lu --- mm/page_alloc.c | 15 +++++++++++++++ 1 file changed, 15 insertions(+) diff --git a/mm/page_alloc.c b/mm/page_alloc.c index dafdcdec9c1f..1d838041931e 100644 --- a/mm/page_alloc.c +++ b/mm/page_alloc.c @@ -1141,6 +1141,9 @@ static void free_pcppages_bulk(struct zone *zone, int count, batch_free = count; do { + unsigned long pfn, buddy_pfn; + struct page *buddy; + page = list_last_entry(list, struct page, lru); /* must delete to avoid corrupting pcp list */ list_del(&page->lru); @@ -1150,6 +1153,18 @@ static void free_pcppages_bulk(struct zone *zone, int count, continue; list_add_tail(&page->lru, &head); + + /* + * We are going to put the page back to the global + * pool, prefetch its buddy to speed up later access + * under zone->lock. It is believed the overhead of + * calculating buddy_pfn here can be offset by reduced + * memory latency later. + */ + pfn = page_to_pfn(page); + buddy_pfn = __find_buddy_pfn(pfn, 0); + buddy = page + (buddy_pfn - pfn); + prefetch(buddy); } while (--count && --batch_free && !list_empty(list)); } -- 2.14.3 From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mail-pg0-f69.google.com (mail-pg0-f69.google.com [74.125.83.69]) by kanga.kvack.org (Postfix) with ESMTP id DBB696B0007 for ; Thu, 1 Mar 2018 01:27:57 -0500 (EST) Received: by mail-pg0-f69.google.com with SMTP id l14so2198177pgn.21 for ; Wed, 28 Feb 2018 22:27:57 -0800 (PST) Received: from mga12.intel.com (mga12.intel.com. [192.55.52.136]) by mx.google.com with ESMTPS id b60-v6si2536498plc.830.2018.02.28.22.27.56 for (version=TLS1_2 cipher=ECDHE-RSA-AES128-GCM-SHA256 bits=128/128); Wed, 28 Feb 2018 22:27:56 -0800 (PST) From: Aaron Lu Subject: [PATCH v4 3/3] mm/free_pcppages_bulk: prefetch buddy while not holding lock Date: Thu, 1 Mar 2018 14:28:45 +0800 Message-Id: <20180301062845.26038-4-aaron.lu@intel.com> In-Reply-To: <20180301062845.26038-1-aaron.lu@intel.com> References: <20180301062845.26038-1-aaron.lu@intel.com> Sender: owner-linux-mm@kvack.org List-ID: To: linux-mm@kvack.org, linux-kernel@vger.kernel.org Cc: Andrew Morton , Huang Ying , Dave Hansen , Kemi Wang , Tim Chen , Andi Kleen , Michal Hocko , Vlastimil Babka , Mel Gorman , Matthew Wilcox , David Rientjes When a page is freed back to the global pool, its buddy will be checked to see if it's possible to do a merge. This requires accessing buddy's page structure and that access could take a long time if it's cache cold. This patch adds a prefetch to the to-be-freed page's buddy outside of zone->lock in hope of accessing buddy's page structure later under zone->lock will be faster. Since we *always* do buddy merging and check an order-0 page's buddy to try to merge it when it goes into the main allocator, the cacheline will always come in, i.e. the prefetched data will never be unused. In the meantime, there are two concerns: 1 the prefetch could potentially evict existing cachelines, especially for L1D cache since it is not huge; 2 there is some additional instruction overhead, namely calculating buddy pfn twice. For 1, it's hard to say, this microbenchmark though shows good result but the actual benefit of this patch will be workload/CPU dependant; For 2, since the calculation is a XOR on two local variables, it's expected in many cases that cycles spent will be offset by reduced memory latency later. This is especially true for NUMA machines where multiple CPUs are contending on zone->lock and the most time consuming part under zone->lock is the wait of 'struct page' cacheline of the to-be-freed pages and their buddies. Test with will-it-scale/page_fault1 full load: kernel Broadwell(2S) Skylake(2S) Broadwell(4S) Skylake(4S) v4.16-rc2+ 9034215 7971818 13667135 15677465 patch2/3 9536374 +5.6% 8314710 +4.3% 14070408 +3.0% 16675866 +6.4% this patch 10338868 +8.4% 8544477 +2.8% 14839808 +5.5% 17155464 +2.9% Note: this patch's performance improvement percent is against patch2/3. [changelog stole from Dave Hansen and Mel Gorman's comments] https://lkml.org/lkml/2018/1/24/551 Suggested-by: Ying Huang Signed-off-by: Aaron Lu --- mm/page_alloc.c | 15 +++++++++++++++ 1 file changed, 15 insertions(+) diff --git a/mm/page_alloc.c b/mm/page_alloc.c index dafdcdec9c1f..1d838041931e 100644 --- a/mm/page_alloc.c +++ b/mm/page_alloc.c @@ -1141,6 +1141,9 @@ static void free_pcppages_bulk(struct zone *zone, int count, batch_free = count; do { + unsigned long pfn, buddy_pfn; + struct page *buddy; + page = list_last_entry(list, struct page, lru); /* must delete to avoid corrupting pcp list */ list_del(&page->lru); @@ -1150,6 +1153,18 @@ static void free_pcppages_bulk(struct zone *zone, int count, continue; list_add_tail(&page->lru, &head); + + /* + * We are going to put the page back to the global + * pool, prefetch its buddy to speed up later access + * under zone->lock. It is believed the overhead of + * calculating buddy_pfn here can be offset by reduced + * memory latency later. + */ + pfn = page_to_pfn(page); + buddy_pfn = __find_buddy_pfn(pfn, 0); + buddy = page + (buddy_pfn - pfn); + prefetch(buddy); } while (--count && --batch_free && !list_empty(list)); } -- 2.14.3 -- To unsubscribe, send a message with 'unsubscribe linux-mm' in the body to majordomo@kvack.org. For more info on Linux MM, see: http://www.linux-mm.org/ . Don't email: email@kvack.org